Modeling Threshold Effects of Information Processes in Social Networks with Induced Atypical Group Behavior

This paper has been devoted to the development of a methodology for predictive analysis of transient processes in network structures with induced group behavior. Interpreting phase portraits of information process models is essential for predicting collective behavior activation scenarios. Destructive activity aimed at targeting group behavior is based on social media’s ability to promote instantly topics in mention rankings. Disturbing cohesive social communities is actively used in planning information attacks and fraudulent activities. Impulsive ripple effects on target groups lead to online epidemics and the spread of aggressive virus-like content. The author of this paper has repeatedly encountered campaigns promoting aggressively attention-grabbing fake information with the destructive goal of promoting vaccination refusal and the use of unacceptable medications during the recent pandemic. Similar posts and comments do not formally violate social media rules and are not direct advertising or threats, but they are used to manage social change. Simulation modeling of adaptive network processes in the social engineering of subcultural groups is a new contemporary problem. Identifying and blocking bot accounts of fictitious personalities (bot farms and troll factories) that specifically control chains of atypical network activity is difficult, but necessary. Complex socially significant phenomena (economic collapses, stock market crashes, resource depletion, invasive outbreaks, hype waves, panic attacks, and fake news) are accompanied by abrupt threshold transitions for previously seemingly stable processes. We conduct a theoretical classification of unstable states and extreme situations with explosive activity observed in reality. We call the entire possible set of such nonsmooth phenomena permissible for the studied process a nonlinear set of situation scenarios. The problem of situational forecasting and modeling of disturbance waves in the information space of social networks is becoming relevant. The method of trigger functionals makes it possible to identify specific trends in situations of parallel tension growth in contact networks as prognostic indicators and predictive factors for the explosive activation of “hype waves.” The author develops original approaches to computer simulation modeling of extreme events and crisis social and biophysical processes to predict situations where multiple interrelated nonlinear effects arise. Many of the model effects are merely redundant properties of the mathematical apparatus. Discrete models of impulse propagation through the network information space are used for situational forecasting and scenario analysis, but the impact of changes in their parameters is sometimes too great. The main feature of the proposed methods is the idea of delimiting the model parameter intervals for interrelated nonlinear effects and bifurcations, which avoids unnecessary chaos. A method that allows for the description of critical and unstable states in models in the presence of oscillatory modes in another region of attraction that simulate the transition to a series of waves has been proposed. Thus, we were able to identify signs of natural surges and waves of excitement around an important issue and the distinctive properties of artificially inducing network storms using fake accounts and botnets.